Is #unstub being called at any point?
Scott
On Feb 16, 2011, at 12:54 AM, Fearless Fool wrote:
> I'm looking at an example where a stub seems to work sometimes, and
> sometimes appears to become "unstubbed". I haven't boiled it down to a
> minimal example, but it goes something like this:
> --
Hmm - when I print puts(premise) rather than puts(premise.inspect), I
can see that they actually have different addresses, so they're not
really equal:
just stubbed #
entering lookup_stuff_on_the_web with #
Now my problem is figuring out why/where the other Premise is getting
created.
In
I'm looking at an example where a stub seems to work sometimes, and
sometimes appears to become "unstubbed". I haven't boiled it down to a
minimal example, but it goes something like this:
the model:
class Premise << ActiveRecord::Base
def lookup_stuff_on_the_web
$stderr.puts("entering
I keep getting an issue when running RSpec with Autotest. My
~/.autotest file looks like this:
==FILE BEGINS==
require 'autotest/growl'
require 'autotest/fsevent'
Autotest.add_hook :initialize do |autotest|
autotest.add_mapping(/^spec\/requests\/.*_spec\.rb$/) do
autotest.files_matc
On Feb 15, 11:14 am, Christoph Schiessl wrote:
> Thanks for your suggestion Justin, but I don't believe that the problem is
> time zone related. Time objects usually don't "loose" their Time Zone when
> performing operations on them. Here's an example for illustration:
>
> $ rails console
> Lo
Thanks for your suggestion Justin, but I don't believe that the problem is time
zone related. Time objects usually don't "loose" their Time Zone when
performing operations on them. Here's an example for illustration:
$ rails console
Loading development environment (Rails 3.0.4)
ruby-1.8.7-p330 :
Your expectation (should_receive) is expecting "start_of_day", which
uses Time.zone. The actual "on_day" scope does
"day.to_time.beginning_of_day", which does not use any time zone.
Therefore, the arguments to in_interval are not the same as the
expectation. And because they are not the same, the m
Hi!
I'm trying to test the following (simplified) model:
class Allocation < ActiveRecord::Base
scope :in_interval, (proc do |start_of_interval, end_of_interval|
params = {:s => start_of_interval, :e => end_of_interval}
where("(starts_at > :s AND starts_at < :e) OR (ends_at > :s AND ends
I'm using ree via rvm. Everything was just installed.
In autotest, the tests won't even start to run. When I run rspec
spec, the tests will run, but will crash at some point here:
src/tcmalloc.cc:390] Attempt to free invalid pointer: 0x10022d9a0
I'm not really sure where to look. Any recommend
On 2011-02-15 14:56:52 +0200, David Chelimsky said:
On Feb 15, 2011, at 9:11 AM, Andrei Maxim wrote:
I could not find a way to tell RSpec to not load the
ActionMailer-related modules and classes. Is there a setting in the
configuration that will do that? Or is this a bug?
This was fixed in
On Feb 15, 2011, at 9:11 AM, Andrei Maxim wrote:
> Hi all,
>
> I'm working on Rails 3 application with ActiveRecord and ActionMailer
> stripped out (I've commented out the respective lines in
> config/application.rb), but I'm running into some problems when I'm running
> the specs either via t
Hi all,
I'm working on Rails 3 application with ActiveRecord and ActionMailer
stripped out (I've commented out the respective lines in
config/application.rb), but I'm running into some problems when I'm
running the specs either via the `rake spec` command or by running
`autotest`.
I could n
12 matches
Mail list logo
